INDEX

Explanations

phrases indicating personal experiences or actions

oai_token-act-pair · gpt-3.5-turbo

verbs indicating actions taken by the speaker

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

wayne

-0.68

 Individuals

-0.64

vation

-0.64

Â¯Â¯Â¯Â¯

-0.64

aic

-0.63

ween

-0.62

bia

-0.62

Entry

-0.62

Known

-0.61

bies

-0.60

POSITIVE LOGITS

 myself

1.00

EStream

0.79

é¾

0.77

eah

0.67

cyclopedia

0.66

 sugg

0.66

 pict

0.66

undai

0.65

 looph

0.65

cffff

0.63

Activations Density 0.290%

phrases indicating personal experiences or actions

verbs indicating actions taken by the speaker

No Comments

No Known Activations