INDEX

Explanations

words related to reasoning or justification

oai_token-act-pair · gpt-3.5-turbo

words or phrases indicating a sense of reasonableness or sufficiency

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

 ways

-0.71

raltar

-0.69

gery

-0.69

 fixture

-0.68

ger

-0.67

ials

-0.67

 methods

-0.66

uese

-0.62

 substitution

-0.62

glers

-0.62

POSITIVE LOGITS

 situated

0.91

 priced

0.82

 spaced

0.82

 positioned

0.81

beit

0.79

nerg

0.78

 entertained

0.77

greg

0.76

 equipped

0.76

 suited

0.75

Activations Density 0.033%

words related to reasoning or justification

words or phrases indicating a sense of reasonableness or sufficiency

No Comments

No Known Activations