INDEX

Explanations

phrases or terms related to processing and managing information

oai_token-act-pair · gpt-3.5-turbo

references to information processing or data-related concepts

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

 Parables

-0.75

atos

-0.75

jug

-0.75

kus

-0.72

alone

-0.72

awar

-0.71

gg

-0.70

versions

-0.66

aughs

-0.63

ny

-0.63

POSITIVE LOGITS

 overload

0.93

 retrieval

0.88

 glean

0.88

frame

0.82

 theoret

0.81

 gathered

0.80

 conveyed

0.77

 gathering

0.77

ãĥĨ

0.77

 Gathering

0.76

Activations Density 0.040%

phrases or terms related to processing and managing information

references to information processing or data-related concepts

No Comments

No Known Activations