INDEX

Explanations

terms related to interception or intercepting, especially in the context of communications and military actions

oai_token-act-pair · gpt-3.5-turbo

terms related to interception in a military or communication context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Aid

-0.71

fred

-0.66

Graph

-0.66

 Legends

-0.65

itsch

-0.64

enny

-0.63

cil

-0.63

pet

-0.63

bold

-0.62

adders

-0.62

POSITIVE LOGITS

 intercept

0.99

 intercepted

0.96

ibles

0.84

ezvous

0.79

ors

0.76

ãĥ¼ãĥĨãĤ£

0.74

 eaves

0.72

 interception

0.72

IBLE

0.71

elligence

0.71

Activations Density 0.018%

terms related to interception or intercepting, especially in the context of communications and military actions

terms related to interception in a military or communication context

No Comments

No Known Activations