INDEX

Explanations

words related to visual stimuli and actions

oai_token-act-pair · gpt-3.5-turbo

dynamic visual elements and actions in descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

chery

-0.90

icians

-0.82

saf

-0.82

hers

-0.80

itism

-0.79

paying

-0.79

TRY

-0.78

thodox

-0.77

ches

-0.76

aligned

-0.76

POSITIVE LOGITS

 Archdemon

0.85

 sands

0.84

 sickness

0.80

 tide

0.79

 masses

0.73

 tides

0.72

sun

0.72

 fortunes

0.71

 hearts

0.70

NESS

0.70

Activations Density 0.148%

words related to visual stimuli and actions

dynamic visual elements and actions in descriptions

No Comments

No Known Activations