INDEX

Explanations

conditional statements suggesting potential outcomes

oai_token-act-pair · gpt-3.5-turbo

conditional phrases expressing hypothetical outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Named

-0.68

 Writing

-0.64

 Chal

-0.61

 Returning

-0.59

 dwelling

-0.59

 Trap

-0.58

charged

-0.57

 Xuan

-0.57

sho

-0.57

 Moving

-0.56

POSITIVE LOGITS

be

0.98

 ideally

0.96

 undoubtedly

0.96

 surely

0.95

 imply

0.94

 suffice

0.94

 probably

0.93

 entail

0.92

 doubtless

0.91

 allow

0.90

Activations Density 0.148%

conditional statements suggesting potential outcomes

conditional phrases expressing hypothetical outcomes

No Comments

No Known Activations