INDEX

Explanations

words related to negative impacts or changes, such as reductions, decreases, or declines

oai_token-act-pair · gpt-3.5-turbo

terms related to decreases or reductions in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

gets

-0.64

flix

-0.63

odore

-0.58

rones

-0.53

 Hancock

-0.52

mates

-0.52

 disse

-0.51

 vain

-0.50

got

-0.50

VICE

-0.50

POSITIVE LOGITS

of

0.99

luster

0.93

 thereof

0.90

pour

0.82

in

0.79

issan

0.77

gradient

0.76

OF

0.74

lust

0.74

 therein

0.71

Activations Density 0.176%

words related to negative impacts or changes, such as reductions, decreases, or declines

terms related to decreases or reductions in various contexts

No Comments

No Known Activations